class: center, middle, inverse, title-slide # SIVOCS --- <style> .center2 { margin: 0; position: absolute; top: 50%; left: 50%; -ms-transform: translate(-50%, -50%); transform: translate(-50%, -50%); } .large { font-size: 130% } .small { font-size: 70% } .remark-slide-content.hljs-default { border-top: 60px solid #23373B; } .remark-slide-content > h1 { font-size: 30px; margin-top: -75px; } </style> ## B1: How familiar are you with the concept of “social innovation” -- **H**: The familiarity with the concept of SI depends on the field of research. --- # B1: Is SI familiarity normally distributed? .pull-left[ ``` ## ## Shapiro-Wilk normality test ## ## data: data$familiarWithSI.response. ## W = 0.87186, p-value < 2.2e-16 ``` * H_o = Normally distributed ==> SI familiarity is not normally distributed ] .pull-right[ <img src="data:image/png;base64,#16_survey_tests_files/figure-html/unnamed-chunk-2-1.svg" width="864" /> ] --- # B1 (OLD) : ANOVA test: SI familiarity depends on sci. domains? .small[ Caution, we are assuming familiarity with SI has an interval scale and variance homogenity] ``` ## Df Sum Sq Mean Sq F value Pr(>F) ## domain 2 451.4 225.70 28.76 2.62e-12 *** ## Residuals 357 2801.5 7.85 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## 1 observation deleted due to missingness ``` ``` ## # A tibble: 3 × 3 ## group1 group2 p.value ## <chr> <chr> <dbl> ## 1 Humanities and Social Sciences Biology and Medicine 6.22e-11 ## 2 Math., Natur. and Eng. Sci. Biology and Medicine 2.65e- 1 ## 3 Math., Natur. and Eng. Sci. Humanities and Social Sciences 8.26e- 9 ``` --- # B1 (NEW): Kruskal-Wallis test: SI familiarity depends on sci. domains? ``` ## ## Bartlett test of homogeneity of variances ## ## data: familiarWithSI.response. by domain ## Bartlett's K-squared = 12.931, df = 2, p-value = 0.001556 ``` ``` ## ## Kruskal-Wallis rank sum test ## ## data: familiarWithSI.response. by domain ## Kruskal-Wallis chi-squared = 45.694, df = 2, p-value = 1.196e-10 ``` * Mean ranks of the groups are not the same, SI fam. depends on domain --- # B1: Wilcox test: SI familiarity depends on sci. domains? .pull-left[ * SSH vs other ``` ## ## Wilcoxon rank sum test with continuity correction ## ## data: familiarWithSI.response. by SSH ## W = 8590, p-value = 2.001e-11 ## alternative hypothesis: true location shift is not equal to 0 ## 95 percent confidence interval: ## -3.000023 -1.999999 ## sample estimates: ## difference in location ## -2.000026 ``` ] .pull-right[ * NS vs other ``` ## ## Wilcoxon rank sum test with continuity correction ## ## data: familiarWithSI.response. by Math._Natur._and_Eng. ## W = 17358, p-value = 0.003771 ## alternative hypothesis: true location shift is not equal to 0 ## 95 percent confidence interval: ## 0.0000604398 1.0000123644 ## sample estimates: ## difference in location ## 0.9999553 ``` ] --- # B1: *Familiarity with SI* across scientific domains <br> <img src="data:image/png;base64,#16_survey_tests_files/figure-html/unnamed-chunk-10-1.svg" width="864" style="display: block; margin: auto;" /> <br> <br> * *Familiarity with SI* differs across scientific domains (ANOVA test: p < 0.05) * *Biology and Medicine* and *Math., Natur, and Eng. Sci.* are similar (pairwise t-test: p > 0.05) * *Humanities and Social Sciences* are significantly different than the others (pairwise t-test with each: p < 0.05) --- class: clear, center **H**: Familiarity with SI correlates with familiarity with transdisciplinarity --- # C2: Correlation with SI familiarity .pull-left[ ``` ## ## Spearman's rank correlation rho ## ## data: b1domain.df$trans_exp and b1domain.df$familiarWithSI ## S = 4454962, p-value = 4.957e-14 ## alternative hypothesis: true rho is not equal to 0 ## sample estimates: ## rho ## 0.3871266 ``` ] .pull-right[
] --- class:clear, center # A2 **H**: Transdisciplinary research depends on age --- # A2: Fam. with Trans. normal distributed? ``` ## ## Shapiro-Wilk normality test ## ## data: data$transdisciplinaryExp.rate. ## W = 0.94158, p-value = 1.487e-10 ``` * Reject H_o = A2 normal distributed --- # A2: Transdisciplinary research depends on age ? * Homogeneity of variances ``` ## ## Bartlett test of homogeneity of variances ## ## data: transdisciplinaryExp.rate. by age ## Bartlett's K-squared = 1.5345, df = 4, p-value = 0.8205 ``` ``` ## ## Kruskal-Wallis rank sum test ## ## data: transdisciplinaryExp.rate. by age ## Kruskal-Wallis chi-squared = 2.4474, df = 4, p-value = 0.6541 ``` * Can't reject the H_o = there is no stat. significant difference between age groups in terms of transdisciplinary experience --- class:clear # D1: Motivation types H: Motivation to improve the human condition/welfare (D1.c) correlates with contribution to better services/products for general population. .pull-left[ ``` ## ## Spearman's rank correlation rho ## ## data: data.questions$motivation.welfare. and data.questions$impactTargetGroup.pub. ## S = 3282110, p-value < 2.2e-16 ## alternative hypothesis: true rho is not equal to 0 ## sample estimates: ## rho ## 0.5406923 ``` ] .pull-right[
] --- class:clear ## H: Generating deeper/better understanding of a specific social issue depends on (the level of) transdisciplinary involvement of citizens ``` ## ## Shapiro-Wilk normality test ## ## data: data.questions$Impactstatements.understanding. ## W = 0.81298, p-value < 2.2e-16 ``` ``` ## ## Shapiro-Wilk normality test ## ## data: data.questions$groupsInvolved.citiz. ## W = 0.5649, p-value < 2.2e-16 ``` --- ``` ## ## Kruskal-Wallis rank sum test ## ## data: Impactstatements.understanding. by groupsInvolved.citiz. ## Kruskal-Wallis chi-squared = 44.018, df = 2, p-value = 2.765e-10 ``` * Different involvement levels make difference. --- ``` ## ## Spearman's rank correlation rho ## ## data: data.questions$Impactstatements.understanding. and data.questions$groupsInvolved.citiz. ## S = 2207337, p-value = 1.404e-11 ## alternative hypothesis: true rho is not equal to 0 ## sample estimates: ## rho ## 0.3901644 ``` ---
--- # D1: Motivation Types ## H: Motivation Types depend on sci. domains --- ###### Phenomenon .pull-left[ <img src="data:image/png;base64,#16_survey_tests_files/figure-html/unnamed-chunk-23-1.svg" width="504" /> ] .pull-right[ ``` ## ## Shapiro-Wilk normality test ## ## data: data.questions$motivation.pheno. ## W = 0.65462, p-value < 2.2e-16 ``` ``` ## ## Kruskal-Wallis rank sum test ## ## data: motivation.pheno. by domain ## Kruskal-Wallis chi-squared = 0.037102, df = 2, p-value = 0.9816 ``` * No stat. sig. difference between domains ] --- ###### Problem .pull-left[ <img src="data:image/png;base64,#16_survey_tests_files/figure-html/unnamed-chunk-26-1.svg" width="504" /> ] .pull-right[ ``` ## ## Shapiro-Wilk normality test ## ## data: data.questions$motivation.prob. ## W = 0.87279, p-value < 2.2e-16 ``` ``` ## ## Kruskal-Wallis rank sum test ## ## data: motivation.prob. by domain ## Kruskal-Wallis chi-squared = 8.3372, df = 2, p-value = 0.01547 ``` * Stat. sig. difference between domains ] --- ###### Welfare .pull-left[ <img src="data:image/png;base64,#16_survey_tests_files/figure-html/unnamed-chunk-29-1.svg" width="504" /> ] .pull-right[ ``` ## ## Shapiro-Wilk normality test ## ## data: data.questions$motivation.welfare. ## W = 0.91547, p-value = 3.131e-13 ``` ``` ## ## Kruskal-Wallis rank sum test ## ## data: motivation.welfare. by domain ## Kruskal-Wallis chi-squared = 31.592, df = 2, p-value = 1.38e-07 ``` * Stat. sig. difference between domains ]